Next: Kalman Filtering in Chandra Aspect Determination
Up: Software Applications
Previous: The ST5000: An Attitude Determination System with Low-Bandwidth Digital Imaging
Table of Contents -
Subject Index -
Author Index -
Search -
PS reprint -
PDF reprint
Petreshock, J. G., Wolk, S. J., Takashi, I, & Cresitello-Dittmar, M. 2001, in ASP Conf. Ser., Vol. 238, Astronomical Data Analysis Software and Systems X, eds. F. R. Harnden, Jr., F. A. Primini, & H. E. Payne (San Francisco: ASP), 467
An Object Oriented Design for Monitoring the Chandra Science Instrument X-ray Background
J. G. Petreshock, S. J. Wolk, M. Cresitello-Dittmar, T. Isobe
Harvard-Smithsonian Center for Astrophysics,
60 Garden Street, Cambridge, MA 02138
Abstract:
The Monitoring and Trends Analysis (M&TA) System for the Chandra X-ray
Observatory consists of multiple software threads designed to monitor and
visualize spacecraft behavior. The Science Instrument (SI) background
monitoring is one such thread that is designed to compile a temporally
and spatially ordered table of the observed flux and energy spectrum in
detector coordinates. In this paper we describe the design of the
tools, and applications of the data products generated, and the output
product flexibility.
As part of the M&TA
System (Wolk et al. 2000) the SI Background (SIB) monitoring thread is based on
C++ tools designed with object oriented design
methodologies that provide flexibility for generating SIB maps. These tools and several
PERL wrapper scripts are woven together to form a pipeline (see Figure 1) used by the
Chandra
X-ray Center Data Systems (CXCDS) pipeline processing system (Plummer 2001). The resultant products
will provide a means for visualization and quantitative analysis of temporal variations in sky
emission and instrument background.
The extremely high angular resolution and sensitivity of Chandra makes this a
unique problem among active X-ray telescopes. But these extremes also hold the promise of
producing a background map of unparalleled resolution. We will discuss and demonstrate the
current suite of tools used for monitoring the SIB for a single Chandra Flight
Instrument, the Advanced CCD Imaging Spectrometer (ACIS).
Understanding Chandra's background is an important factor in driving Science
Operations (SciOps) decisions. For example, if we detect lower
background levels while the spacecraft is within the magnetotail then we can schedule
observations of low surface brightness targets during these times.
Figure 1:
SIB pipeline flow chart.
 |
Depending upon the SI configuration and individual observation configurations,
Chandra can collect well over 1.0GB worth of event data per
week. This large data volume was another aspect of the SIB monitoring problem that had to
be handled. We solved this problem by binning the data by time according to
event time stamps, thus increasing data manageability.
The SIB monitor is designed to detect temporal and spatial variations
in the background count rates and characterize flare properties including: rates, intensities,
and durations. Flares are identified by comparing background count rates for each energy
band against predefined limits. Background values that exceed these
limits are flagged as flares. This is demonstrated by Figure 2, which diagrams a
one dimensional view of the parameterized temporal and energy binning performed by the SIB
tools. These parameters are controlled via the individual tools
parameter file using the CXCDS parameter interface library.
The flexibility of the parameterization provides
control over the data volume and thus increased data manageability.
This flexibility also allows users to create products that range from a purely
instrumental background analysis to a finely binned sky map.
The SIB monitoring generates FITS compliant binary tables.
The CXCDS DataModel (DM) provides the interface between
the classes and the FITS file reading and generation. Each class is equipped
with data members that store pointers to DM variables used to connect with
the FITS file.
Figure 2:
Single energy band Binning Diagram.
 |
The five classes created for the SIB monitor were designed to
interface using cascading levels of data encapsulation. This design
provides flexibility with class handling. With only slight
modifications, specifically function overloading, these classes were
capable of being used within both the mta_gen_sibkg tool that
generates the SIB monitor products and the mta_merge_sibkg tool
that is used to merge multiple SIB monitor products together. To reduce the memory
usage by the SIB tools, the classes are
designed with time being the primary processing key, followed
by sky coordinates (Right Ascension, and Declination) and finally energy. X-ray Events
from input files are processed per time bin (see Figure 2).
When an event from a new time bin is detected, the objects flush
their data to a FITS file and the object containers are reset in
preparation for receiving data from the new time bin. The SIB classes are
described in the following subsections in order from base
classes to higher level derived classes.
This class is the basis of the M&TA SIB monitoring, the EnergyBand (EB)
class stores information about energy bands such as the number of events,
energy ranges, the band name, and a pointer to the output FITS column.
The EB class functions to store the information for a time bin and is
responsible for writing the data to FITS file.
The EnergyBandContainer is a container class for EB
objects and stores the information for the energy spectrum as defined
by the user configurable energy band definitions files. This
container class provides a clean interface hiding the work of event
processing from the higher level classes.
The SkyLocation class is a derived class from EnergyBandContainer. A
SkyLocation object describes the energy spectrum for a given parcel of the sky.
For the case that the sky coordinates are undersampled (i.e., sky is resolved
into one pixel) the SkyLocation object represents the SIB.
The mtaSIBkgMap is a container class for SkyLocation objects
and represents the full observed background map. This is the class
responsible for controlling the highest level of event processing and
provides a clean control interface with the lower level objects.
This class is used to read and store the EB definitions indicated
by the energy band definitions parameter. The sole function for this class it to
provide a template to populate new EnergyBandContainer objects. This
class is not a derived class, and operates in conjunction with the
mtaSkyBkgMap class to initialize the required objects for event processing.
The SIB monitor products for the ACIS instrument contain a column for each
energy band defined within the input energy band definitions files as
well as a time column and a flare column. The flare column is
defined as an N element bit array that represents each energy band.
These FITS products are currently used to generate light curves, flare
statistics, and temporal spectral variability plots.
In any experiment, understanding the signal and background components
within a data set is a crucial step, and SIB monitoring provides a useful
quantitative analysis of several background properties. The analysis
of SIB properties can also be used to update target scheduling,
to increase Chandra's observational efficiency.
Our application of these simple classes allows us to handle
the large volume of X-ray data from Chandra efficiently and
effectively. The SIB monitor products are ideal for integration into
the M&TA trend analysis thread and for use in multisystem correlations with data
in existing M&TA databases. The rebinning of sky coordinates allows a
basic means for generating the observed X-ray background
map. With future versions of the M&TA SIB monitoring we plan to
provide the capability
to generate higher resolution background maps by applying archived aspect solutions to the
background information in detector coordinates. Also, the knowledge acquired by monitoring
Chandra's SIB can be applied toward future missions and thus potentially increase
cost effectiveness.
Acknowledgments
This project is supported by the Chandra X-ray Center under NASA contract NAS8-39073.
References
Petreshock, J. G., Wolk, S. J., & Cresitello-Dittmar, M. 2000, in ASP Conf. Ser., Vol. 216, Astronomical Data
Analysis Software and Systems IX, ed. N. Manset, C. Veillet, &
D. Crabtree (San Francisco: ASP), 475
Plummer, D. 2001, this volume, 475
Wolk, S. J., et al. 2000, in ASP Conf. Ser., Vol. 216, Astronomical Data
Analysis Software and Systems IX, ed. N. Manset, C. Veillet, &
D. Crabtree (San Francisco: ASP), 453
© Copyright 2001 Astronomical Society of the Pacific, 390 Ashton Avenue, San Francisco, California 94112, USA
Next: Kalman Filtering in Chandra Aspect Determination
Up: Software Applications
Previous: The ST5000: An Attitude Determination System with Low-Bandwidth Digital Imaging
Table of Contents -
Subject Index -
Author Index -
Search -
PS reprint -
PDF reprint
adass-editors@head-cfa.harvard.edu